BEYOND MANDE MORY: Islam and Ethnicity in Côte d’Ivoire

Robert Launay Northwestern University

Marie Miran

Colby College

(article paru en juillet 2000 dans Paideuma, n° 46 : 63-84) The categorization of African peoples was a necessary adjunct to its colonial annexation by Europeans. This is hardly surprising if one accepts Foucault’s assertion that “... power and knowledge directly imply one another;... there is no power relation without the correlative constitution of a field of knowledge, nor any knowledge that does not presuppose and constitute at the same time power relations” (Foucault 1979: 27). Specifically, in the case of the French (and no doubt of other colonial powers), among the first intellectual tasks of the colonial administration was the classification of Africans in terms of ethnicity and religion. Africans were divided into discrete “races”, a term which, in French, does not necessarily carry as aggressively racist a connotation as in English (Amselle and M’Bokolo 1985); at the same time, they were categorized as “fétichistes” with various beliefs and practices, or as Muslims, and furthermore as followers of one or another Sufi brotherhood (Harrison 1988). At first glance, it might seem that the logics of classification by ethnicity and by religion are diamterically opposed. Ethnicity is, in principle, relatively fixed and hereditary. Religion, on the other hand, is relatively fluid and potentially a matter of personal choice. As far as the French authorities were concerned, this was particularly evident in the case of the “fetishist” majority, potential converts to Christianity but also, of course, to Islam. Moreover, the very processes whereby colonial administrators arrived at these classifications seem to differ radically. Ethnicity itself, it has been argued, was very much a construction of colonial systems of knowledge (Amselle and M’Bokolo 1985; Amselle 1990). This is not to suggest by any means that ethnic categories were entirely a figment of the colonial imagination, but rather that, out of a welter of disparate and sometimes contradictory markers of identity and of distinction, colonial authorities chose to privilege certain ones and to ignore others entirely in order to construct a mosaic of discrete and exclusive “ethnies”. The very act of naming was often a colonial gesture if only because it was often the case that, as one of the leading scholar/administrators of the early colonial period lamented in a monographic study of “the Siéna or Sénoufo people”, “the people in question do not always seem to acknowledge a generic name for themselves” (Delafosse 1908-09: 17). It is far harder to argue that Islam per se was a colonial construction. The selfsame Africans who might not be able to identify themselves as “Sénoufo” were quite capable of inidicating whether or not they were Muslim. The real power of these colonial categories of knowlege lay perhaps less in their administrative utility-- the order they created was arguably more conceptual

2

than political -- nor necessarily as a discourse legitimating the exercise of colonial authority, akin to Orientalism (Said 1978); rather, it was most radically demonstrated by the rapidity with which Africans themselves came to take this system of classification for granted. In this respect, ethnic and religious categories, however disparate their origins, came to represent the primary idioms of identity construction and of community formation. However, precisely to the extent that Africans identified themselves as members of religious and of ethnic communities, these two principles of identity entered into a dialectic relationship with one another. This is to say that the relationship between “ethnicity” and “religion” is an issue that only made sense in terms of colonial systems of knowledge, but which assumed a dynamic of its own which eluded the control of the colonial, and later the postcolonial, state. The question has raised itself with particular acuity in Côte d’Ivoire, if only because of the remarkable diversity of the colony (and later the nation) in ecological, linguistic, cultural, and religious terms. The country is roughly bisected by an ecological frontier, the southern half consisting of rain forest while the northern half is wooded savanna. The area is home to approximately sixty different languages from at least four different major language families. Broadly speaking, the ecological north-south divide is paralleled by a lingustic divide, with Akan and Kru languages spoken in the south and Mande and Voltaic languages in the north. (To make matters more complicated, the Mande family is divided into “southern” and “northern” branches. Southern Mande languages -- for example Yacouba, Gouro, and Beng -- are spoken in the south of the country. Throughout the rest of the article, we will use the term “Mande” to refer excusively to northern Mande speakers, who acknowledge a common origin in the land of Manden -- in fact, the medieval empire of Mali.) Cultural diversity, while it certainly does not correspond in precise ways to linguistic diverstiy, is certainly as extensive. As for religion, aside from the obviously very disparate array of beliefs and practices which were concealed by colonial rubrics such as “fetishism” or “animism”, the French found on their arrival an important Muslim community living as a religious minority throughout much of the northern zone. Moreover, early on in the colonial period, Christianity began to make significant inroads in the southern part of the country, not only in the form of Roman Catholicism but also of mainstream Protestant denominations and indeed independent African Churches. We propose in the remainder of this essay to examine in some detail the particular relationship between Islam and ethnicity in Côte d’Ivoire. At the outset of colonial rule, this relationship depended on the articulation of prior systems of social organization, particularly in the northern, savanna zone, with the categories which French colonial administrators imposed on social realities as a means of understanding and ruling subject populations. Specifically, Islam was associated with specific ethnicities, those the French termed “Malinke” and “Dioula”. However, such categories, initially creations of French colonialism, took on a dynamic of their own independent of direct colonial control. Specifically, we will show how a qualitatively new kind of “Dioula” identity centered on new urban centers in the southern part of the country, while preserving if not enshrining the link between Mande ethnicity and Islam, reconfigured the relationship in very

3

substantial ways in the course of the colonial period. Finally, we will discuss how and why, in postcolonial Côte d’Ivoire, a new generation of Muslims is actively and self-consciously seeking to disarticulate this link between Islam and ethnicity. Islam and the Mande in Côte d’Ivoire In an early survey of Islam in the relatively newly created colony of Côte d’Ivoire, Paul Marty estimated that about one-sixteenth of the total population of the colony -- 100,000 out of 1,600,000 inhabitants -- was Muslim, the remainder being “animiste ou fétichiste.” In fact, Marty asserted, “la famille mandé ... renferme tous les musulmans,” “the Mande family encompasses all the Muslims.” Admittedly, there were exceptions: several dozens (“quelques douzaines”) of Islamized Senufo and Kulango who, by the virtue of their very conversion -- or so Marty suggested -- were in the process of becoming “denationalized” (Marty 1922:5). Elsewhere in his survey, he mentions a number of other non-Mande Muslims, for example, a number of Baoule converts in the village of Aoussoukro (Marty 1922: 52), not to mention a variety of individuals he labels “Maures”, a category which, for Marty, seems to encompass all native Arabic speakers, not only from Mauritania but also from Algeria, Tunisia, and even the Hijaz (Marty 1922.: 69-76). Such exceptions notwithstanding, it was clear that, at the outset of the twentieth century, almost all the Muslims in Côte d’Ivoire were speakers of Mande. The converse, however, was by no means the case: according to Marty, only one of three Mande speakers was Muslim (Marty 1922: 5). In spite of the “Mande-ness” of Islam in Côte d’Ivoire (at least in the eyes of colonial beholders), the relationship between language, religion, and identity was by no means straightforward. First of all, there was considerable dialectical variation in the Mande spoken throughout the colony; in some cases, different dialects of Mande were mutually incomprehensible. This is particularly true of the dialects spoken in the southern stretches of the savanna towards the west of Côte d’Ivoire: mau kan, spoken around Touba; worodu kan around Séguéla; and konyara kan around Mankono (Person 1968:52). On the other hand, the fact that Mande was the lingua franca of trade throughout the Ivoirian savanna -- the entire northern half of the colony -- militated in favor of a certain degree of standardization, of the preservation of an mutually intelligible core alongside regional dialectical variation. The trade networks which had for centuries crisscrossed northern Côte d’Ivoire, extending well beyond its borders, were characterized both by the practice of Islam as a religion and the use of Mande as a language -- hence the link between the language and the religion so apparent to Marty. On the other hand, the majority of Mande speakers were not involved to any great extent as participants in these trading networks, and so were far less likely to categorize themselves as Muslims. Specific lineages, often bearing certain dyamuw -- Mande patronyms -- were known for specializing in trade and/or Islamic scholarship. Even now, the praise songs of griots perpetuate the memory of such specialization: “Cisse ye mande mory ye, Toure ye mande mory ye”, “The Cisse (or Toure) are “mande mory”. In its most restricted sense, the word mory (or mori) refers to an Islamic scholar, and individual whose religious learning entitles him to authority in that domain. Certain dyamuw

4

were renowned for their scholars: the Saganogo and Diane, for example. Much more generally, mory were all those persons who, by their hereditary membership in certain lineages, were expected to conform rigorously to Sunni standards of piety: regular prayer five times daily, fasting during the month of Ramadan, abstinence from forbidden foods and alcoholic beverages, etc. Such standards of piety were the hallmark of Muslim identity in the trading networks of which many Mande were integrated; specialized trading lineages, as well as scholarly ones, were generally of mory status. In the northwest, where Mande speakers (grouped by the French under the rubric “Malinke”) constituted the overwhelming majority of the population, mory status and Muslim identity tended to coincide. Towards the central and eastern parts of northern Côte d’Ivoire, where Mande (referred to as “Dioula” or “Mande-Dioula” by the French) lived as a minority among a variety of other populations -- Senufo, Kulango, Abron -- most if not all Mande considered themselves Muslims, but not necessarily mory; lineages of “warrior” status, called “tun tigi” in Korhogo or “sonongui” in Kong, also usually assumed a Muslim identity without conforming to mory norms of religious practice (Launay 1982: 23-47; Green 1986). Since mory status was hereditary, it was entirely possible for individuals from other lineages to adopt similar religious practices, without becoming mory themselves; only if the entire lineage changed its religious practices did it become (or for that matter cease to be) mory. In short, at the outset of the colonial period, almost all Muslims in Côte d’Ivoire were Mande, but the majority of Mande were not Muslims; Muslim identity was by and large hereditary, and, even within the Muslim community, religious practice varied according to hereditary membership in lineages of different status. Islam and “Dioula” ethnicity under colonial rule Entirely unintentionally -- in large measure, in contradistinction to the stated aims of colonial policy -- colonial rule was to transform radically the relationship between Islam and ethnic identity. Before the colonial period, Mande speakers had no attachment to a common identity as such, even if they spoke a common language and acknowledged a common (but distant) origin in the land of Manden. On one hand, individuals were subjects of particular states or chiefdoms, not all of which were ruled by Muslims or for that matter by Mande speakers. On the other, mory were conscious of belonging to a global network of Muslim traders and clerics which transcended linguistic boundaries. Thus, the great Mande trading towns of Kong and Bondoukou each had its marrabaso, its Hausa quarter, as clearly shown on maps of each town drawn by Binger (1892: vol. 1, 294; vol. 2, 167). The French acknowledged (or invented) a variety of Mande-speaking ethnic groups: Bambara, Marka, Malinke, Dioula (only the latter two being “native” to Côte d’Ivoire). However, the dynamics of ethnicity did not depend exclusively on the French labeling process. Colonial rule allowed individuals a considerably greater freedom of movement, especially throughout Côte d’Ivoire but also more generally throughout the French territories of l’AOF (Afrique Occidentale Française) and even to territories like the Gold Coast (now Ghana), under the control of other colonial powers. Specifically, the whole southern half of Côte d’Ivoire, much of which lies

5

within the forest zone, was opened up to “northerners”. Before the colonial period, the frontier had in fact been jealously guarded by residents on both sides (Launay 1978), who shared a common interest in monopolizing access to scarce goods, notably kola nuts. The opening of this southern frontier was by no means immediate. The pacification of parts of southern Côte d’Ivoire was not complete by the outbreak of the First World War (Angoulvant 1916; Weiskel 1980). However, as soon as it was safe to travel, Mande-speaking communities sprouted in the forest zone, close to the sources of kola, as entrepots in the long-distance trade, effectively breaking the long-established monopolies of the towns along the frontier. Even more than kola nuts, coffee and cocoa were to make the southern part of the country a pole for migration from the north. In the first place, the plantations required seasonal wage labor, recruited not only in northern Côte d’Ivoire but, perhaps even more, in neighboring colonies, notably Mali and Haute-Volta (now Burkina Faso). Ultimately, many northerners were able to purchase rights over arable land and to establish plantations in their own right, at the cost of incurring considerable resentment on the part of the local populations (Dupire 1960: Dozon 1985: 276-302). More generally, the prosperity of southern Côte d’Ivoire throughout much of the colonial period and after, compared to the “underdeveloped” north -- both within and outside the colony -- generated the growth of large and medium sized towns, not to mention the megalopolis that Abidjan was eventually to become. These towns also constituted important poles of migration. Malians, Voltaics, and peoples from northern Côte d’Ivoire were particularly active in what is now labeled the “informal sector”: wholesale as well as petty trade; transportation; weaving and tailoring. A very substantial proportion (though by no means all) of these “northern” migrants were native speakers of Mande. This was particularly the case for Malians, although Peul and Songhay speakers were also active migrants. There were also a considerable number of Mande speakers from Haute Volta, especially from the eastern part of the country around Bobo-Dioulasso, which was indeed attached to Côte d’Ivoire during part of the colonial era; however, the majority of Voltaic migrants were speakers of More (Mossi) or other related Voltaic languages. Mande speakers also predominated among migrants from northern Côte d’Ivoire, although there were a considerable numbers of Senufo, speaking a Voltaic language (quite different from More). Guinean migrants as well, though not as numerous, were often Mande speakers. Around this core of Mande speakers there developed a new identity -- “Dioula”. The term itself had long existed in Mande, both to refer to professional traders and to Mande-speaking minorities in parts of Côte d’Ivoire, but it came to have a qualitatively different meaning in the south of the country (Launay 1982: 106-22). Typically, “Dioula” were Mande speakers, Muslims, and active in trade (specifically) or the informal sector (more generally). This association of language, religion, and economy was to have important consequences. The movement of “northerners”, “Dioula” or otherwise, into southern Côte d’Ivoire was accelerated by the advent of modern means of transport: first the railroad, extending northwards from Abidjan and eventually reaching Ouagadougou; and, after the Second World War, by the diffusion of modern motor transportation and the construction of a

6

network of roads throughout the colony. By the 1950s if not earlier, a substantial proportion of the populations of virtually every Mande-speaking community in northern Côte d’Ivoire was living in the south. For example, a survey of 144 married males in the village of Kadioha in north-central Côte d’Ivoire conducted in 1973 found that nearly 40% of them were living in other towns of the country (about 23% in the south); that nearly 70% had spent some time as migrants elsewhere (the vast majority for over five years) and that no less than 34% had lived abroad for 10 or more years (Launay 1982: 98-99). While one cannot extrapolate such figures too literally for other communities -- much less for twenty years or so beforehand -- they indicate an order of magnitude which was, to judge from accounts of life histories of Kadioha residents, no recent phenomenon. Migrants such as the men and (to a somewhat lesser extent) the women of Kadioha and other such villages were easily integrated into the “Dioula” communities of southern Côte d’Ivoire. Such integration was virtually obligatory for anyone who wanted to enter the informal trading sector. More to the point, becoming “Dioula” also entailed becoming “Muslim”, if one was not already so. This “Dioual-ization” of Mande speakers in southern Côte d’Ivoire led to the virtually total Islamization of Ivoirian Mande, not only in towns but also in the rural areas.1 This Islamization of Mande speakers spread, not only throughout northern Côte d’Ivoire, but throughout those colonies, most notably Mali and Haute Volta, whose denizens migrated in massive numbers to their more prosperous southern neighbor. Indeed, the Bambara or Bamana, whose very name supposedly denoted their rejection of Islam, were overwhelmingly Muslim by the end of the colonial period. The emergence of a “Dioula” identity went hand in hand with the development of a more or less homogeneous “Dioula” culture. Language and religion were linchpins of this urban culture. Mande as spoken in the rural areas of northern Côte d’Ivoire was subject to considerable dialectical variation. A more or less standardized “Dioula-kan” began to emerge in urban areas as the dominant lingua franca of the marketplace, indeed as one of the three hegemonic languages of Côte d’Ivoire, along with French (the only official language) and -- to a much lesser extent -- Baoule. French was the language de rigueur in the formal sector: schools, hospitals, colonial administration, etc. But Dioula-kan predominated in the informal sector, and a large proportion of the Ivoirian population, even southerners, were able to speak at least some Dioula, if only to do their shopping. A similar kind of homogenization operated in the religious domain. Islam, as practiced by the urban “Dioula”, involved the relatively strict application of mory standards of piety regarding regular prayer, fasting, alms, and abstention from alcohol and impure meats. This was the practice fully expected of the growing ranks of new converts. The kind of double religious standard which had once applied in certain Mande speaking communities, where individuals in different hereditary categories were held accountable to different patterns of religious observance, was now felt to be highly inappropriate, not only in towns but ultimately in the home 1 The causes and consequences of this massive Islamization during the colonial period, not only among the Mande but more generally in French colonial West Africa, are discussed in more detail in Launay and Soares (in press).

7

communities (Launay 1992). Young boys were routinely sent to study with religious teachers, where they received a standard education aimed principally at teaching them to recite (indeed to psalmody; such recitations were inevitably melodic, if hardly invariably harmonious) prayers and passages from the Qur’an. Funeral ceremonies were, in particular, a vehicle for the articulation of a standardized religious culture. If weddings were, to an important extent, a family affair, conducted under the aegis of local traditions, funerals were markers of identity in the broader Muslim community, the umma. In the first place, proper procedures for washing the corpse, reciting prayers over it, and its inhumation are strictly prescribed in standard Maliki manuals of Islam law, such as the Risala of Ibn Abi Zayd al-Qayrawani (1968:105-15). Above and beyond these requirements, there developed the standard practice of performing “recitations” (kalan), virtually sermons, forty days after the burial, as well as a cycle of ritual prestations -- classified as saraka, “alms” -- at various moments of the funeral cycle. Funerals -- particularly sermons -- were in fact a form of recreation in the “Dioula” community. A good preacher was also a rousing orator. However, funerals were a form of sociability most appropriate to older men and women in the “Dioula” community. Adolescents -- kambelenw (young men) and sunguruw (young women) -- were more likely to participate in dance associations (tonw). Such neighborhood associations were, like funerals, instrumental in the elaboration of a standardized “Dioula” culture throughout all the towns of Côte d’Ivoire and which, through the movement of migrants to and from the north to the south, spread as well to rural Mande-speaking areas. Indeed, it was not obligatory for urban “Dioula” to come from Mande-speaking backgrounds. Large numbers of Senufo from north-central Côte d’Ivoire who migrated to the south also converted to Islam, became “Dioula” and adopted the use of Dioula-kan in public (not necessarily at home). Most strikingly, such Islamized and Dioula-ized converts not only adopted, as might be expected, Islamic proper names -- Amadou, Bakari, Fatoumata, or Salimata, for example; they also adopted Mande patronyms such as Coulibaly or Kone, reinforcing the symbolic identification of Islam as a religion and Mande ethnicity. This urban “Dioula” culture did not, by any means, entirely eradicate local particularities if not particularisms within the broader “Dioula” community (LeBlanc 1996).2 In the first place, the urban “Dioula” communities were subdivided into smaller communities sharing a common home of origin: Odienne-ka, “people of Odienne; Korhogo-ka, “people of Korhogo”, etc. The -ka suffix, it is important to note, is intrinsically relative. In the town of Korhogo, a native of a nearby village such as Waraniene would be labeled as Waraniene-ka; in the towns 2 In her description of “Dioula” of Malian origin in the town of Bouake, LeBlanc goes as far as to suggest that “Dioula” identity is largely salient in terms of national identity politics in Côte d’Ivoire, largely as a marker distinguishing “northerners” from the locally (and, politically speaking, nationally) predominant Baoule, whereas “ethno-cultural” differences characterize Malians from different communities of origin in their home country. However, her conclusion that “l’identité dioula ... n’est pas associée à une base de pratiques culturelles spécifiques (LeBlanc 1996: 72)” underrates the extent to which the construction of a relatively homogeneous linguistic and religious community constituted a significant historical break with past practices, one which has arguably been so successful as to be rendered relatively invisible and eclipsed by cultural differences.

8

of the south, however, the same individual would be “from Korhogo”. In this manner, the broader “Dioula” community was comprised of a limited number of regional subgroupings. These subgroupings functioned as networks of support as well as of power bases, in struggles for the control of the leadership of mosques, as well as for various economic or political resources.3 Cultural differences, for example in marriage practices and ceremonies, could serve as markers of the divisions within the “Dioula” community (LeBlanc 1996), just as funerals marked their cultural and religious unity. Like the -ka suffix, these cultural differences could be used as symbolic markers for differences of varying orders and degrees: e.g., between Dioula from the Odienne region as opposed to Korhogo; between people from Korhogo town and surrounding villages; between people from different neighborhoods in Korhogo town; indeed, between members of different lineages in the same neighborhood in Korhogo. Perhaps the most fundamental difference, and certainly the one with the greatest political significance, was between “Dioula” who considered themselves fully Ivoirian and those who hailed from neighboring countries -- Mali, Guinea or Haute-Volta. The attitude of Ivoirians towards these and other foreigners has long been profoundly ambivalent. On one hand, immigrant labor is essential to the Ivoirian economy. On the other, the country has seen repeated -- sometimes violent -- waves of xenophobia, such as the riots directed against Togolese and Dahomeans in 1958 (Zolberg 1969: 245-48) or, much more recently, against Ghanaians. For some southerners, all Mande speakers, all “Dioula”, were not only “northerners” but, for all intents and purposes, “foreigners”, whatever their origin. Ivoirian “Dioula”, however, had an interest in stressing their status of “natives” fully entitled to a share in the country’s relative prosperity. On the other hand, the informal sector, particularly in the domains of trade and transportation, depended on maintaining trans-colonial (and later trans-national) networks linking Ivoirian and foreign “Dioula”. In short, an individual’s identification with one region of origin or another outside southern Côte d’Ivoire was an integral component of Muslim “Dioula” identity. Such identification cut both ways. On one hand, it tended to structure internal divisions -- and sometimes divisiveness -- within the broader umbrella of the “Dioula” community. On the other, it ensured that “Dioula” however long they had been living in the south (and increasingly, in the case of “Dioula” born in the south, all of their lives), identified themselves and were identified by others as “northerners”. This identification of Ivoirian Dioula living in the south with their “home” communities in the north, was, as we shall see, to have important political consequences in the postcolonial era. In some instances, cultural conservatives of one stamp or another might fight rearguard action against the assimilation of their own people into the “Dioula” cultural amalgam. Launay witnessed parents in Korhogo who would upbraid their children about the “proper” way to speak, insisting that they maintain the use of local dialectical particularities rather than speaking like the “riffraff” on the street. The likelihood, however, that such admonitions would receive a sympathetic ear 3 See Lewis 1970 for a detailed analysis of the internal divisions within the “Dioula” community of a large southern town.

9

among adolescents was minimal, to say the least. In a different vein, certain Senufo intellectuals writing in the 1970s and 1980s were harshly critical of fellow Senufo who adopted “Dioula” patronyms, converted to Islam, and more generally conformed to “Dioula” cultural norms, using colonial ethnography to depict the “Dioula” as literally antithetical to the “Senoufo” (Coulibaly 1978; Tuho 1984; see also Launay 1999). Paradoxically, these very intellectuals had used Western-style education -- familiarity with the French language, not to mention culture -- as a means of entry into the “formal” sector, whereas a far greater number of their fellow Senufo had instead used the possibility of assimilating “Dioula” culture as a means of access to the informal sector. Despite their numerical predominance, the “Dioula” were by no means the only Muslim group in the south of Côte d’Ivoire. Many of the migrants from Haute-Volta, notably Mossi, were not Mande-speakers, and did not assimilate into the “Dioula” category even though a large proportion of them converted to Islam. Many of these Voltaics were employed in sectors of the economy for which, unlike trade, religious identity was not particularly salient: unskilled wage labor and domestic service. (Conceivably, Muslims would have been at a disadvantage as servants, if they were expected to serve alcoholic beverages or, worse, prepare pork for dinner.) Aside from the “Dioula”, these Voltaics were no doubt the largest, though not necessarily the most conspicuous, category of Muslims. Moors from Mauritania, Wolof from Senegal, Peul from Mali and Haute-Volta, and Hausa from Niger formed visible communities, especially in Abidjan but also in other towns of Côte d’Ivoire. (One of the neighborhoods of Korhogo, for instance, was named “Hausabougou”, “Hausatown”; admittedly, Hausa hardly constituted a majority in that neighborhood, but their presence was, as the name suggests, conspicuous.) In short, the Muslim community of colonial Côte d’Ivoire was, in the first place characterized by a division between the Mande-speaking “Dioula” majority and a minority of “foreigners” -- Voltaics, Senegalese, Peuls, Hausa. The “Dioula”, in turn, were further divided between “native” Ivoirian Mande-speakers and “foreigners” from Mali, Guinea or Haute-Volta. Last but not least, the Ivoirian “Dioula” were subdivided into communities acknowledging different regions of origin. In principle, such divisions ought to have been irrelevant in the domain of religion, especially given the specifically universalist discourse of Islam. In fact, they were often reflected in factional rivalries, specifically over the control of mosques. In the 1930s, the Muslim communities of Agboville, Bouake, Daloa, Dimbokro, Gagnoa and Man witnessed factional struggles over the position of imam. These divisions almost always corresponded to differences in ethnic or regional origins. For example, in Gagnoa in 1934, a quarrel broke out concerning the appearance of the new moon marking the beginning of Ramadan, revealing smoldering tensions between Odienneka and Worodouka (from the Worodougou region around Mankono.) The Worodouka refused to follow the lead of the Odienneka, one of whom held to office of imam. The quarrel turned violent, and several rioters were jailed by the colonial authorities. As a result, the Worodouka decided to split off from the rest of the Muslim community and to build a new mosque. Eventually, through the mediation of an outside religious leader (Seydou Nourou Tall, a Senegalese leader of the Tijaniyya and a direct descendant of al Hajj

10

Umar Tall, but also a “marabout” with very close ties to the French colonial authorities), a compromise was reached whereby the Worodougouka would pray separately during the week in their own mosque but join the Odienneka in the Friday mosque. During the very same period, in the Treichville neighborhood of Abidjan, another quarrel broke out in 1931 along different lines, revolving around the construction of a new mosque. The old mosque, built early in the 1920s, whose imam was Fetigue Konate, a “Dioula” from Haute Volta, came to be known as the “Dioula mosque”. The new mosque was built by a group of Senegalese notables, Wolof for the most part though also Toucouleur, with a Senegalese imam, Thierno Souley Sall. Not surprisingly, it came to be known, first as the “Wolof mosque” and later as the “Senegalese mosque”. It was the intention of the builders of the new mosque that it would be the sole locus of prayer for the Muslim community every Friday, with the older mosque relegated to the status of secondary mosque reserved for daily prayers. However, the “Dioula” community refused to follow the lead of the Senegalese and continued to pray under the leadership of a “Dioula” imam. Once more, Seydou Nourou Tall proposed his services as mediator, initially in 1931 and again in 1933. Although he was successful in eventually defusing the situation, the Treichville neighborhood continued to host not one but two distinct Friday mosques, each with its own “ethnic” appellation. It is important, on the other hand, to avoid exaggerating the importance of these cleavages within the Muslim community. Not all Muslim communities in Côte d’Ivoire were characterized by factional splits by any means. In Abidjan, with the largest Muslim community and consequently the largest proliferation of mosques, there were admittedly yet other mosques -- e.g., the “Peul mosque” -- associated with ethnic or national labels. But most mosques were known (informally) by the location of local landmarks: the “Texaco mosque”; the “Abattoir [slaughterhouse] mosque”; the “St. Michel mosque” (near the St. Michel church); and even the “Bracodi Bar mosque”. Most often, convenience rather than ethnicity or, for that matter, such factors as Sufi brotherhood affiliation dictated the believer’s choice as to where to pray, punctuated by occasional factional crises which tended to polarize the community along ethnic lines. This was, in fact, what happened at the “Senegalese Mosque”. By the end of the 1930s, less than a decade after its controversial founding, it boasted a multi-ethnic congregation, including not only Senegalese and foreign as well as Ivoirian “Dioula”. Indeed, when another conflict over leadership broke out in 1939, the imam at the time, Souley Sall, was Tukulor. When Sall’s leadership was contested by Wolof notables, he received the backing of “Dioula” as well as of Tukulor at the mosque; and when the Wolof temporarily prevailed and he was destituted of office, Tukulor and “Dioula” deserted the “Senegalese mosque” for the “Dioula mosque” of Treichville, at least in the short run. In the long run, though, the congregation of the “Senegalese mosque” -- like those of the other mosques of the capital -- came to be ethnically mixed, with a large “Dioula” contingent reflecting, after all, the overall composition of the Muslim community in the capital, not to mention in the colony at large. Post-colonial politics: severing the ethnic link

11

The “Dioula-ization” and Islamization of large numbers of Ivoirians and of immigrants from neighboring countries was initially a response to changing economic conditions, to the opening up of an “informal” sector throughout the forest zone of Côte d’Ivoire. However, in the context of the newly independent Republic of Côte d’Ivoire, Islam and ethnicity were to play a very different role within the context of national identity politics. For its small size, Côte d’Ivoire remains remarkably diverse ecologically, linguistically, culturally and religiously. Upon independence, no single group enjoyed a majority of any sort. The one-party state which Houphouët-Boigny established and ruled until his death in 1993 was predicated on the maintenance of a delicate balance whereby all major groups in the country had -- admittedly in quite varying degrees -- some stake in the system. Broadly speaking, the complex ethnic mosaic (bearing in mind that these various “ethnies” were not the mechanical reflections of real cultural diversity but rather a product of the colonial process) crystallized into five major regional blocs: the North (“Dioula” and “Senufo”); the Center (Baoule); the East (Agni); the West (Bete); and the Coast (“lagunaires”, lagoon cultures, including Ebrie, Alladian, and Aboure among others). Houphouët-Boigny was himself from the Center, and it was commonly acknowledged that the Center received a disproportionate share of the “spoils” of the system. The location of the capital, Abidjan, on the Coast also privileged that region. (The location of the capital was also eventually moved to Houphouët-Boigny’s birthplace, Yamoussoukro, in the Center; even so, Abidjan remains the economic center of the country, and the Coast has thus not entirely lost its advantage.) Both East and West were the major coffee and cocoa producing regions in the country, the basis of its prosperity during the economic boom period -- the so-called (and short-lived) “Ivoirian miracle” of the first decade after Independence. However, for this very reason, a sizable proportion of the land on which these plantations were situated had been acquired by “outsiders” from the region -- “Dioula” but also “Baoule” among others. As a result, separatist movements emerged in both the East in 1959, just before Independence, and in the West in 1970 (Zolberg 1969: 292-93; Dozon 1985: 344-46); both movements were rapidly and forcefully suppressed. The North was, by contrast, distinctly “underdeveloped”, both economically and educationally. Within this regional system, the situation of the “Dioula” was in some respects anomalous. More than any other group, they were scattered throughout the country. Despite their large numbers in the towns of the south, and the length of time many of them had been living there, they were nonetheless identified with their region of “origin”, not with their place of residence. Moreover, the “Dioula” were, as we have seen, divided into Ivoirians -- those who had a real stake in the North of the country, even if they were not living there at the time -- and foreigners, with “home” ties outside Côte d’Ivoire. The educational backwardness of the North contributed to the political disadvantage of “Dioula” as well as of other northerners. As late as 1962-63, only 17% of school age children in the northern prefecture of Korhogo were enrolled in primary school, as opposed to an average of 45% for the country as a whole and rates such as 77% for Dimbokro, 82% for Abengourou, and 97% for Agboville, all

12

in the south of the country (S.E.D.E.S. 1965, vol. 1: 60). The boom years of the 1960s saw a massive expansion of the formal sector, particularly government employment, as the personnel of schools and other government services were gradually Africanized. Salaries were comparatively high, pay was relatively regular and predictable, and there were additional “perks” including family allowances and subsidized (in some cases free) housing. Foreign “Dioula” were virtually excluded from such positions, but there were relatively few Ivoirian “Dioula” with the educational credentials to take advantage of the opportunities, and so the formal sector was overwhelmingly dominated by southerners. Very crudely and imprecisely, one might characterize Côte d’Ivoire in the 1960s in terms of a largely Christian (especially Catholic) formal sector, a Muslim informal sector, and an “animist” (the term adopted by the Ivoirian census) peasantry. Indeed, religions, like regions, could constitute blocs to compete with more or less success for government resources. Although the construction of the vast majority of churches and mosques was financed locally and informally, the government would also from time to time undertake the building of showpiece edifices as a gesture to one or another religious community. The government would also sponsor national youth conferences of one religion or another, and, perhaps most important of all, could allocate television time for religious programming, no doubt the most dramatic and effective way for any religious community to signal its “national” importance. Given Houphouët-Boigny’s very public allegiance to the Roman Catholic Church, Catholics were -- certainly in the eyes of practitioners of other religions -- accorded more than their proportionate share. The construction of an imposing Cathedral in Abidjan and, in the final years of Houphouët’s reign, the massive basilica of Yamoussoukro (financed, Houphouët was to insist, with his own personal funds, but skeptics alleged that the President never made a very firm distinction between national and personal resources) were certainly the most spectacular of these showpiece edifices. This competition for resources among religious groups had important consequences, not only for the respective national visibility of Catholics, Protestants, and Muslims (“animists” were, in the absence of any institutional structures, virtually excluded from the game), but also for leadership within the Muslim community. In the first place, most forms of government largesse had to be administered by one formal organization or another, in the name of the Muslim community as a whole. Mosques were admittedly an exception, becoming, as it were, the collective property of the local Muslim community as a whole. However, conferences had to be organized, religious television programs to be produced, activities which an explicitly secular government could not undertake on its own initiative. In the absence of any institutionalized Church which might serve as a channel for such sponsorship, Muslims were at a distinct disadvantage, particularly after 1963 when the government, allegedly in order to foil an attempted coup, placed a ban on all formal associations. The lifting of the ban in the 1970s opened the door for the creation of formal Muslim associations, and in 1979, the CSI (Conseil Supéieur Islamique, Islamic Superior Council) was formed as an umbrella for Muslim associations throughout the country. The CSI thus aspired to act as a channel for resources from the government as well as from outside donors, not least

13

among them conservative Arab nations such as Saudi Arabia, Kuwait, and the Gulf Emirates. However, the very extent to which the CSI was politically well connected to the ruling government rendered it suspect to the majority of the Muslim community, As a result, other Muslim organizations sprang up, most notably the AEEMCI (Associations des Eleves et Etudiants Musulmans de Côte d’Ivoire, the Muslim Students’ Assocation), with the explicit stance of maintaining a respectable distance from the ruling party. The profile of the leadership of these new associations differs sharply from that of the traditional leadership of the “Dioula” Muslim community. In fact, the mory scholarly lineages who had dominated the field of Islamic learning before the colonial period had continued to exercise their hegemony over the newly constituted “Dioula” communities of southern Côte d’Ivoire. A religious leader’s credentials depended on his isnad, his scholarly pedigree stating the identity of his teacher, his teacher’s teacher, and so forth (Wilks 1968; Launay 1990). This is not to say that the old scholarly lineages maintained a strict monopoly on advanced religious learning, but rather that they remained its guarantors, those through whom a proper pedigree had to be traced. Moreover, they embodied a particular style of learning, one which emphasized rote memorization; to know a text meant, in the first instance, to know it by heart. For the new leadership, training in the Arab-speaking world -- Egypt or Saudi Arabia, for example -- and the capacity to speak, read and write “Arab” Arabic was the hallmark of religious authority. This new generation of religious leaders differed radically from the elders, not as much in content as in style, far more depersonalized and disembodied. In fact, organizations such as the AEEMCI have been remarkably successful in attracting a younger generation of Muslims, especially those who are in one way or another disenchanted with the old style of mory leadership. Muslims who had been educated in state-run Western-style schools were among the first to experience this disenchantment. It is a paradox that, at the very moment that the Ivoirian economy was experiencing a sharp downturn that sharply affected all those employed in the formal sector, who saw their perks and salaries stagnate if not shrink, the educational lag separating the north from the south was, if by no means eradicated, at least seriously reduced. Proportionately more and more Muslims were employed by the formal sector in what were admittedly less and less attractive positions. At the same time, declining prices on the world market adversely affected coffee and cocoa planters, so that, by the 1980s, the differences in economic prosperity as well as in education between north and south, Muslims and Christians, were, though by no means leveled, certainly far less acute than they had been at the outset of Independence. But for this new generation of Western-educated Muslims, the old style of mory scholarship was something of an embarrassment, with its emphasis on rote learning and melodic recitation. Ultimately, education was the terrain whereby the new generation of Arab-trained Muslim scholars was to prevail over their mory predecessors. In part with financial and technical assistance from the Arab world, particularly from Saudi Arabia, they were able to establish a new type of Islamic school, the madrasa, which could successfully compete both with old style Qur’anic education (which it was virtually to supplant in its entirety in a remarkably short period of time) as well

14

as with the state-run Western-style school system. The madrasa was very much a hybrid. In terms of curriculum, its emphasis on religious values and on the Arabic language contrasted sharply with the emphatically secular state school system. On the other hand, the organization of instruction resembled that of the state schools, with classrooms, desks, blackboards, textbooks, worksheets, and indeed,, even in the instruction of Arabic, an emphasis on grammar and vocabulary (and in general an emphasis on literacy skills), as opposed to the rote memorization of sacred texts. Indeed, some of the madrasas also taught subjects in the state curriculum preparing students for the national Primary School Certificate exam. This compromise often fell short of the ideal, if largely because the Ivoirian government refused to acknowledge any diplomas awarded by such schools. In the absence of any possibility of government accreditation, there was no control over the quality of instruction, and some madrasas fell woefully short of their promises. Moreover, it was impossible to transfer from the Islamic to the state system, even in the case of students who successfully passed the Primary School Certificate exam; admission to public secondary schools required an official transcript from a recognized institution, which the madrasas were unable to furnish. However, as economic prospects deteriorated dramatically in the formal sector, especially in government service, such a handicap was less likely to matter to many Muslim parents and students. For graduates of the madrasas as well as of the state system, the younger Arab-trained scholars corresponded far better to their expectations than did the older generation of mory clerics. For them, too, learning meant the abstract and relatively depersonalized mastery of “skills” and “values”, rather than the personalized bond of apprenticeship between teacher and disciple at the very core of the mory conception of scholarship. This radical change in the style as well as in the personnel of Islamic leadership has had radical effects in restructuring the relationship between Islam and “ethnic” identity. Allegiance to the mory scholars had been a cornerstone of the formation of a “Dioula” identity in the towns of the Ivoirian south. It was they who presided over the collective public ritual manifestations of simultaneous “Dioula” and “Muslim” identity -- prayer at the Friday mosque and the performance of proper Muslim funerals. For the new generation, on the contrary, the mory scholars represented the embodiment of a “Dioula-ness” from which they had good reason to distance themselves.4 “Dioula” identity had always entailed a subsidiary identity in terms of community of origin, entailing a host of moral obligations, not only to fellow “hometowners” in the southern cities, but also to family members back home, often in rural areas. The new generation of students, most often born in the cities, were not necessarily very familiar at all with their putative “home” communities, and, as they grew older and financially independent, more and more reluctant to acknowledge moral -- especially financial-- claims on their strained resources on the part of their rural relatives or even relatively distant urban kinsmen. If a previous generation, eager to enter the informal sector of the prosperous towns of the Ivoirian south, had become 4 The switch from “Dioula” to a more general and generic Muslim identity among youth of Malian origin in Bouake is discussed in great detail and with considerable sensitivity by LeBlanc (1998a, 1998b) to whom we are deeply indebted for our argument.

15

Muslim in order to become “Dioula”, this new generation was equally solicitous of shedding its “Dioula” identity (at least to some degree) while remaining Muslim all the while. The younger generation of Arab-trained Muslim leaders has emphatically echoed this call to reject any “ethnic” dimension in Islam. For example, for the inauguration of the Centre Islamique Bilal in 1992 in a suburb of Abidjan, the preacher, Aboubacar Fofana, insisted: “The mosque belongs to God and to God alone. There is no Dioula mosque, no Senegalese mosque, no Malian mosque, no Koyaga, no Mahou, no Odienneka mosque, but a house of God built by men and women to the glory of their Lord. In this house, there is no place reserved for the privileged. The best places, the first ranks, belong to all Muslims, young or old, who arrive first in the house of Allah.” In an even more radical break with past practice, some of these preachers have been concerned, not only with abolishing “ethnic” divisions within the Muslim community, but also with attracting converts from within those southern “ethnic” communities who have, until recently, remained almost completely impervious to Islam. In 1991, a group of students who graduated from universities in the Arab world (mainly in Saudi Arabia) created LIPCI, the Ligue Islamique des Prédicateurs de Côte d’Ivoire (the Islamic League of Preachers of Côte d’Ivoire). The founding members recruited preacher candidates, noticeably converts, from all over Côte d’Ivoire to preach Islam in local languages in the countryside. By 1992, the LIPCI was claiming 500 preacher-members throughout the country. Its distinctive function is to centralize and coordinate all da’awa (a term which might approximately be translated as “outreach”) activities in the nation. It organizes courses in Arabic, yearly training seminars for preachers and, every two years, training seminars for Qur’anic teachers. It has also initiated mobile Islamic missionary expeditions, called “caravanes de da’awa”, targeting the countryside, with the goal of reaching non-Muslims and supporting isolated Muslim communities.5 These attempts to disembed Islam from “Dioula” identity have taken on new salience in the aftermath of Houphouët’s reign, when the balance of power between regions and between religions in Côte d’Ivoire has been called back into question. In the race to succeed Houphouët, the principal contenders were Henri Konan Bédié -- like Houphouët, a Baoule Catholic from the Center -- and Alassane Outtara, a “Dioula” Muslim from the North. A priori, Outtara was hardly representative either of his “home” region or his religion. A technocrat with a French, Christian wife, he was best known for implementing the IMF’s policy of structural adjustment. However, in a maneuver meant to block his (remote) chances of winning the presidency, the Ivoirian House of Deputies passed a law stipulating that, not only the president, but both his parents had to be native Ivoirians. Since Outtara’s mother hailed from Burkina Faso, he was consequently disqualified. Above and beyond its direct effect, the law implicitly suggested that northerners like Outtara were also, in some measure, really “foreigners” -- a view which, as we have seen, has enjoyed some popularity at certain times in the Ivoirian south. The stratagem was certainly a success in the short run, but, paradoxically, it also had the unintended effect of

5 see Miran 1998 for a fuller discussion of LIPCI and related Islamic movements in Côte d’Ivoire.

16

rallying, not only northerners, but Muslims in general behind Outtara, whose exclusion from power they felt as their own. To the extent that Konan Bédié could successfully play one region off against another, this relative political marginalisation of the north, and in ethnic terms of the “Dioula”, was arguably astute. On the other hand, the demographic balance of power in religious terms has shifted radically in Côte d’Ivoire. At the onset of Independence, no religion could claim a plurality of any sort. However, the last forty years have seen a sharp increase in the proportion of Muslims as well as evangelical Protestants, as opposed to Catholics and “animists”. It is certainly no accident that these are the very religions which have benefited from foreign funds: petrodollars in the case of Islam, American dollars for evangelical Protestantism. It is, of course, far too simplistic to suggest that foreign dollars automatically generate converts. Certainly, in the case of Islam, it is extremely difficult to assess the real extent of these contributions, and whether they represent a substantial influx of funds or rather a symbolic expression of religious solidarity. However, even symbolic gestures such as offering a few scholarships for Ivoirian Muslim students to study in Arab countries may have a considerable impact on the Muslim community’s perception of itself and its relationship to the Islamic world outside the nation’s borders. In any case, Islam can now lay unequivocal claim to the largest religious community in Côte d’Ivoire, and is not very far from becoming the religion of the majority. Consequently, any attempt to marginalise Islam as a religion in favor, notably, of Catholicism would be excessively foolhardy on the part of the Bédié administration. On the contrary, the new regime has, with great fanfare, laid the foundations of a new mosque in Abidjan, situated in the Plateau, the major business district and the economic center, not only of the city but, in many respects, of the country as a whole. (In fact, plans for the construction of the mosque had already been approved before Houphouët’s death, but Konan Bédié, as designated successor, is obviously eager to take the credit.) The Plateau mosque is a gesture to the Muslim community. The publicity which surrounded the beginning of its construction not only represented an attempt to minimize the spill over from the exclusion of Outtara’s candidacy, but also a token that the government fully recognized that Islam shared a central place in the country as a whole. It remains to be seen whether these efforts will eventually persuade the Muslim community, which reacted to the political exclusion of Ouattara as a symbol as a symbol of their exclusion, not only from the political arena, but from the government’s vision of Ivoirian society as a whole. At the same time, Islam, stripped of its “Dioula-ness” which links it too tightly to Mande ethnicity, to the informal sector in the economy, and the north, the most educationally and economically “underdeveloped” region of the country, can far more effectively lay claim in its own right to “national” status. The fact that a younger generation of Muslims, educated in a different system -- whether secular or Islamic -- are increasingly anxious to shed some of their narrowly “northern”, “Dioula” ties lends increasing weight to such a shift in the way in which the Muslim community of Côte d’Ivoire conceives of its place in society and in the political life of the nation.

17

It is still no doubt the case, just as it was at the outset of the century, that most Muslims in Côte d’Ivoire are native speakers of Mande, if only because Dioula, as a lingua franca of youth on the streets of every Ivoirian town, has also, like the religion of Islam, spread well beyond the boundaries of any single “ethnic” community. Such an apparent continuity masks the radical ways in which changing political and economic circumstances have altered the composition of the Muslim community of Côte d’Ivoire and its relationship to “ethnic” and other identities. At the outset, Islam was first and foremost a religion of the mory, that minority of Mande speakers who were active participants in inter-regional trading and clerical networks. The economic development and the rapid urbanization of the Ivoirian south saw the emergence of a new “Dioula” community, in control of the informal sector, whose corporate identity hinged on both a standardized practice of Islam and an equally standardized dialectical version of Mande. Finally, for a new generation of Muslim youth who never experienced colonial rule, but who have had to adjust to the crises accompanying the end of the “Ivoirian miracle” as well as of the Houphouët regime, Islam and ethnicity are now experienced, not as identical but as virtually incompatible. These important shifts in the relationship of Islam to ethnicity in Côte d’Ivoire should caution us against an oversimplistic application of Foucault’s structures about the inseparability of “knowledge” from “power”. Colonial rule very definitely involved the imposition of specific ways of thinking about identity and community among subject African peoples, in terms of constructions of categories of “ethnicity” and indeed “religion”. (Of course, Islam long predated colonial rule; but it was only with the colonial introduction of Christianity through missionary efforts that “religion” came to be understood in terms of competing religious communities with the framework of a single colony or, later, nation.) However, once these categories were firmly established, these ways of thinking about self and community entrenched, they ceased to be (if ever they had been) in control of French authorities. Rather, they provided frameworks within which Africans could actively configure and reconfigure social relationships, independent of if not, in some cases, diametrically opposed to their French rulers, and later the postcolonial state. Such reconfigurations have emerged as responses to specific changes in the economic and/or political domain: the phenomenal growth of the informal sector in the towns of the Ivoirian south in the colonial period; the shrinkage of the formal sector in line with externally imposed “structural adjustment”; the success or failure of postcolonial governments’ attempts to rule by balancing off regional and religious interests. It is essential to insist that these reconfigurations constitute responses, not mechanical reflections. To argue that they are comprehensible only in terms of shifts in the political economy of Côte d’Ivoire is not to assert that they are determined by such shifts. On the contrary, these reconfigurations have resulted from the deliberate actions of specific agents, and especially of young adults. However much it is true that the emergence of new “Dioula” communities in the Ivoirian south may have fostered conversions to Islam among northern migrants, it remains true that these migrants, on their return home, often had to overcome the recalcitrance, and in some cases overtly defy the authority, of their elders. The new Muslim associations, even if their leaders are exceptionally careful to avoid

18

antagonizing oder generations, appeal particularly to the young. If the young play such an active role, this is not because of any age-old “conflict of generations”. Rather, this is because shifts in political economy have very different impacts on members of different generations, opening up or closing off different opportunities. At the same time, as the percentage of the population living in urban as opposed to rural communities continues to grow, a generation which has grown up and lived all its life in the towns of the south is far less likely to feel “at home” in any real sense in their parents’ community of origin. In these ways, within a single century, different generations of young Muslims have come to redefine their own identity in very different terms, not once but twice. Should Muslims become an outright religious majority in Côte d’Ivoire, and particularly if efforts to convert significant numbers of “southerners” begin to bear fruit, then it seems likely that the next generation will, in its turn, redefine Muslim identity in Côte d’Ivoire in a radically new fashion once again.

19

Bibliography AMSELLE, Jean-Loup 1990 Logiques Métisses. Paris: Payot. AMSELLE, Jean-Loup and ETIENNE M’BOKOLO,eds. 1985 Au Coeur de l’Ethnie. Paris: La Découverte. ANGOULVANT, G. 1916 La Pacification de la Côte d’Ivoire. Paris: E. Larose. BINGER, Louis Gustave 1892 Du Niger au Golfe de Guinée, 2 vols. Paris: Hachette. COULIBALY, Sinali 1978 Le Paysan Senoufo. Abidjan and Dakar: Nouvelles Editions Africaines. DELAFOSSE, Maurice 1908-09 “Le peuple siéna ou sénoufo. Revue des Etudes ethnographiques et

sociologiques. I: 16-32; 79-92, 151-59, 242-5, 448-57,483-86; II: 1-21. DOZON, Jean-Pierre 1985 La Société Bété. Paris: Karthala. DUPIRE, M. 1960 “Planteurs autochthones et étrangers en basse Côte d’Ivoire Occientale.”

Etudes Eburnéennes 8: 7-237. FOUCAULT, Michel 1979 Discipline and Punish: The Birth of the Prison. Translatied by Alan

Sheridan. New York: Vintage Books. GREEN, Kathryn 1986 “Dyula and Sonongui Roles in the Islamization of the Region of Kong.

African and Asian Studies 20: 97-117. HARRISON, Christopher 1988 France and Islam in West Africa, 1860-1960. Cambridge: Cambridge

University Press. LAUNAY, Robert 1978 “Transactional Spheres and Inter-Societal Exchange in Ivory Coast.” Cahiers

d’Etudes Africaines 18,4 : 561-73. 1982 Traders Without Trade: Responses to Change in two Dyula Communities.

Cambridge: Cambridge University Press.

20

1990 “Pedigrees and Paradigms: Scholarly Credentials among the Dyula of Northern Côte d’Ivoire.” In Dale Eickelman and James Piscatori, eds., Muslim Travellers. London: Routledge, pp. 175-99.

1992 Beyond the Stream: Islam and Society in a West African Town. Berkeley and Los Angeles: Unifersity of California Press.

1999 “Stereotypic Vision: The “Moral Character” of the Senufo in Colonial and Postcolonial Discourse. Cahiers d’Etudes Africaines, 154, XXXIX-2: 271-92.

LAUNAY, Robert and BENJAMIN SOARES in press “The formation of an ‘Islamic sphere’ in French Colonial West Africa.

Economy and Society LEBLANC Marie Nathalie 1996 “Etre Dioula à Bouake ou comment transiger divers référents identitaires,

Bulletin du Groupement Interdiscipinaire en Sciences Sociales - Côte d’Ivoire, 14: 59-74.

1998a “Between Modernity and Tradition: Rituals of Islam as Negociation and Renewal through Knowledge.” unpublished paper, American Ethnological Association and Canadian Assocation of Social and Cultural Anthroplogy, Toronto.

1998b “From Ethnicity to Islam: Processes of identification amongst Muslim youth in Bouake, Côte d’Ivoire.” unpublished paper, Mande Studies Conference, Banjul.

LEWIS, Barbara 1970 The Transporters’ Association of the Ivory Coast. unpublished Ph.D.

dissertation, Northwestern University. MARTY, Paul 1922 Islam en Côte d’Ivoire. Paris: Ernest Leroux. MIRAN, Miran 1998 “The Revolution of the Da’awa (Call to Islam): Strategies and Practices

within the Muslim Community of Abidjan, Côte d’Ivoire, in the 1990s.” unpublished paper, African Studies Association, Chicago.

PERSON, Yves 1968 Samori: Une Révolution Dyula, vol 1. Dakar: IFAN. al QAYRAWANI, ibn Abi Zayd 1968 La Risala (tr. L. Bercher). Algiers: Editions de l’Armée. S.E.D.E.S. (Société d’Etudes du Développment Economique et Sociale) 1965 Région de Korhogo: Etude de développment économique et sociale, vol 1,

rapport démographique.

21

SAID, Edward. 1978. Orientalism. New York: Random House. TUHO, Charles Valy 1984 J’ai Changé de Nom ... Pourquoi? Abidjan: Nouvelles Editions Africaines. WEISKEL, Timothy C. 1980 French Colonial Rule and the Baule Peoples. Oxford: Clarendon Press. WILKS, Ivor 1968 “The Transmission of Islamic Learning in the Western Sudan.” in J.R.

Goody, ed., Literacy in Traditional Societies. Cambridge: Cambridge University Press, pp. 162-97.

ZOLBERG, Aristide 1969 One-Party Government in the Ivory Coast, revised edition. Princeton:

Princeton University Press.